神经网络用于色谱研究（Ⅰ）──GC保留值估算

Neural Networks Used in Chromatography (Ⅰ)──prediction of GC Retention Parameters

神经网络用于色谱研究（Ⅰ）──ＧＣ保留值估算李志良，八重治，梁本熹，石乐明（湖南大学化学化工系，长沙，４１００８２；日本国立丰桥技术科学大学，中国科学院化工冶金研究所）关键词神经网络，修饰反向传播算法，气相色谱保留值神经网络（ＮＮ）近年来获得突破性进...

Neural networks algorithm was applied to chromatography. The relationship between GC retention parameters and induction effect I, hydrophobic index lgP, molar refractivity MR and molecular connection ￣1χ was studied by using the modified backpropagation (MBP) neural networks. The GC retention indices of alphatic amines, alcohols, acids and esters were estimated and predicted with relative error less than 10%. The neural network method behaved as a good modelling technique for predicting GC retention parameters and might therefore be regarded as a powerful chemometric method.